© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 7738
    Prev
    Next
    INDEX
    Explanations

    technical specifications and details related to mechanical components in vehicles

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    71
    +0.13
    0.6%
    229
    +0.09
    0.4%
    592
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    365
    +0.13
    0.01
    314
    +0.09
    0.00
    284
    +0.09
    0.02
    Negative Logits
    gypt
    -0.70
     Miko
    -0.68
     wedge
    -0.68
    alam
    -0.66
    venants
    -0.65
     Vaughan
    -0.63
    ucl
    -0.62
    ournals
    -0.61
     disproportion
    -0.60
     Rabb
    -0.60
    POSITIVE LOGITS
    dden
    0.87
    ffff
    0.79
    0010
    0.75
    VALUE
    0.75
    Shift
    0.72
    FFFF
    0.71
    ippi
    0.67
    0
    0.66
    200000
    0.66
    00
    0.65
    Activations Density 0.101%

    No Known Activations